I also just tested with PRAGMA journal_mode = OFF; and, while it did prevent sqlite from making and removing journal files all the time, it turns out that it didn't make any significant difference. The measurement was actually 5 minutes slow, considering the variability of the machine, probably means it didn't really make any difference.
It might make sense to think it through a bit more carefully, however. This will turn off all ACID promises for the DB. For creating a new DB from scratch, that's not a big deal. But, are you doing something at the beginning with preserving current ids? If so, then there may be some additional risk where we could lose that mapping if care isn't taken. I have looked through the variety of PRAGMAs that sqlite offers, and I didn't see anything else that would make a big impact. Finally, these timings were don't on a dual CPU, 1GHz machine with 1GB ram, and a random mixture of SATA1 and older UDMA/33 harddrives using LVM2, running Debian testing. Not exactly a screaming machine. Just about any modern machine will do much better. mrc ------------------------------------------------------------------------------ This SF.net email is sponsored by: SourcForge Community SourceForge wants to tell your story. http://p.sf.net/sfu/sf-spreadtheword _______________________________________________ Imdbpy-devel mailing list Imdbpy-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/imdbpy-devel
